Despite the rapid acceleration of gene discovery, our understanding of penetrance and variable expressivity remains limited. The ciliopathies have emerged as a useful model to study these phenomena. An improved knowledge of ciliary composition has led to the identification of numerous disease loci, each of which can contribute causal and modifying alleles. Moreover, an appreciation of the signaling functions of the cilium has allowed the assessment of the in vivo pathogenic potential of alleles. In the previous cycle, we asked how mutational load in the intraflagellar transport (IFT) particle can influence phenotypic variability. Through this work, we identified new disease loci; we showed a significant contribution of CNVs; and we showed how increased IFT mutational load tracks with phenotypic severity. Here, we will extend our studies to understand how a modular approach to genetic disease can improve the predictive power of genetic data, inform genetic buffering mechanisms, and assist the development of therapeutics. Our ultimate goal is to understand how variation across the ciliary module affects expressivity. However, because the entire ciliary proteome remains experimentally intractable, we will anchor our work on robust Preliminary Data gained from the IFT particle and extend to the other two known ciliary particles: the transition zone complex and the BBSome. We propose three Aims. First, through the systematic functional analysis of mutational data, we will ask whether enrichment for mutations in the three complexes tracks with endophenotypes and whether there are genetic interactions between specific subparticles. Second, we have found a source of false negative prediction of pathogenicity to derive from evolutionary conservation, where a human mutant allele is fixed in a different species. Using in silico and in vivo tools, we found that these phenomena can be caused by cis complementation, where a co- evolving allele in the same gene offers a protective effect. We will ask whether this is a common buffering mechanism against deleterious alleles; and whether we can predict cis complementing sites by studying co- evolution within disease-causing proteins and their modules. Finally, we showed recently that proteasome augmentation can rescue core BBSome component defects. However, such manipulations failed to rescue IFT defects, or defects induced by transient BBSome components, suggesting that the position of proteins within modules might have a differential response to therapeutics. We will test this hypothesis by suppressing each component of the three complexes in vivo and in vitro and ask whether genetic or chemical proteasome augmentation can rescue phenotypes differentially. Overall, our studies will determine how the distribution of mutations within modules and between associated modules contribute to variable expressivity; will interrogate the understudied phenomenon of cis-modification, which can inform the pathogenic potential of haplotypes in humans; and will begin to explore how the differential position of mutations in functional complexes can influence response to therapeutic interventions and thus tailor future clinical trials to specific genotypes.